Introduction

• Relational Reinforcement Learning (RRL) paradigm uses relational state (and action) space...
  

Introduction

• The paper introduces a simple data augmentation protocol that provides a...
  

• The paper presents some general ideas and mechanisms for multiple model-based RL. Even...
  

Introduction

• Continual Learning paradigm focuses on learning from a non-stationary stream of...
  

Introduction

• Standard unsupervised learning aims to learn transferable features. The paper proposes...
  

Introduction

• Graph Neural Network (GNN) is a family of powerful machine learning...
  

• Link to the paper

• The paper provides useful empirical advice...
  

Introduction

• The paper presents a framework that uses diverse suboptimal world models...
  

Introduction

• TuckER is a simple, yet powerful linear model that uses Tucker...
  

• Link to the paper

• Training RNNs to model long term...